Glioblastoma (GBM) remains a largely incurable disease, with a 5 year survival rate of less than 10%. Effective targeted therapies to complement already maximal radiation and chemotherapy are urgently needed. Converging evidence shows that the EphA2 receptor is an attractive target for GBM. The PI's lab and others recently uncovered dual roles of EphA2 in tumor etiology and malignant progression. When engaged with ligands (ephrin-As), EphA2 is a tumor suppressor and inhibits both ERK and Akt activities in GBM cells. However, in the absence of ligands, EphA2 is phosphorylated by AGC kinases including Akt and p90-RSK on serine 897, and S897 phosphorylation converts EphA2 from a tumor suppressor into an oncogenic protein that promotes glioma cell migration in vitro and intracranial invasion of glioma stem cells (GSC) in vivo. Moreover level of pS897-EphA2 is correlated with tumor grades. Mechanistically pS897- EphA2 regulates the stem properties of GSCs and promotes gliomagenesis in the absence of ligands. In contrast, upon ligand stimulation, EphA2 induces GSC differentiation and inhibits tumor development. Based on this series of observations, we propose that ligand-mimicking small molecule agonists of EphA2 can be novel therapeutic agents for GBM. Such agonists are expected to i) restore the intrinsic tumor suppressor functions of EphA2, ii) disrupt the pro- oncogenic Akt/RSK-EphA2 signaling axis, and iii) induce differentiation of GSCs. Using structure- guided virtual screening and cell-based assays, we reported that doxazosin (DZ), an ?1- adrenoceptor antagonist still in clinical use for hypertension, is a bona fide EphA2 agonist. DZ inhibits ERK and Akt and suppresses tumor cell dissemination in an EphA2-dependent manner. Much more potent derivatives of DZ have been characterized through medicinal chemistry, including BW27, which suppressed GBM in preclinical models and was capable of crossing the blood-brain barrier (BBB). The overarching goal of this proposal is to translate these basic and preclinical discoveries into nove GBM therapeutic agents. BW27 will be subject to systemic preclinical test across all four molecular subtypes of human GBM. The on-target effects of BW27 will be investigated using the genetically engineered mouse model. Finally the X-ray co-crystal structure of EphA2 in complex with BW27 will be determined to guide the design of next generation(s) of EphA2 agonists. Completion of the studies could lead to new mechanism-based small molecule drug(s) for therapeutic intervention of GBM.